Tolga Oztan

From InterSciWiki
Jump to: navigation, search

Tolga -- Dwight Read

2019 Dec 30

Dear Doug and Lilyan,

I hope all is well with you. I wish you a merry Christmas !! How is life in La Jolla?

I am settling more in Amsterdam - learning some Dutch and moving to a place of my own. I am still at booking.com helping the company with their statistical tests they run on their online platform. I read some interesting books lately - Turchin's Ages of Discord is especially interesting in terms of the predictions he makes for current day US - applying his theory to US as a case study. Also bought Joseph Henrich's The Secret of Our Success' but didn't get the chance to read it yet. But it looks interesting - an entire chapter on in-laws, so I am intrigued.

Hope you are doing great.

Cheers,

-Tolga Joseph Henrich Dec 30 2019

The 2013 AAA Abstract

Susan Rathburn 553 Boz -
File:TolgaUfuk IMG 0104.jpg
left Tolga Ufuk Doug

This is the Washington Univ in St Louis, excellent PostDoc http://pages.wustl.edu/mii Joseph Loewenstein <mii@artsci.wustl.edu>

B.Tolga Oztan.jpg
San Francisco talk AAA 2013

Office: Social Science Tower 949 278 4027 WHW 278 4027

facebook

Sirpsindigi Sokak 
Oztan Apartmani 28/5 
Merter Istanbul 
Turkey

Kinship Simulations - Project Simpa ANR funded - KinshipSimulations.pdf

Generation_permutation_code#Tolga_Oztan_code_Nov_11-11 - Tolga_Oztan_code_Nov_11-11

Tolga_Oztan_code_Node Independent Paths May 2011 - TolgaUfuk_IMG_0104.JPG Tolga Ufuk Doug. TolgaUfuk_IMG_0104.JPG <-- to the right

Tolga Oztan Maps

Tolga Logistic Regressions -- Tolga Logistic Regressions#Plotting avoidance vs. density

April 2017

Patrick McConvell Hi Doug, I saw and downloaded your draft with Oztan on Avoidance/respect (2013). It was due to appear in the Wiley Companion to Cross-cultural Research, but further googling could not find that volume. The draft says I should not cite that but only the (to be) published version.

I have a grant from the Centre of Excellence for the Dynamics of Language, ANU, to run a small conference to run a small conference on the diffusion of f avoidance terminology and practice between Austronesian, Papuan and Australia, and we are waiting for a result on another application to the British Academy to mine some databases on the same topic.

I also gave a paper at the conference on Hunter-Gatherers in Cambridge UK in September and will be submitting a revision shortly. It is about 'demand sharing' in Australia but also talks about trade/gift exchange, particularly an instance where groups used 'silent trade'. Reading your paper with Oztan I am now wondering if this is connected with in-law avoidance. I'd probably like to cite your paper in that context.

I was preparing to edit a Companion of Kinship for Wiley but they dropped the project and the editor left the company. I hope your book is still due to be published, or if not I can cite the paper somehow.

Best,

Pat

Quick Tips for Modifying the Working Codes

 (I used the third one in the list (CreateModelValueSDWchildren2.R) as a template since it had the most number of independent variables. 
 
 Dr. White has detailed, step-by-step instructions for the necessary modifications under Working *Rccs* models, I highly recommend you go over them
 carefully first and then visit my tips if you're running into some problems.
 

First set of changes:

 change the existing dependent variable with your own dependent variable. To do this you change two lines. 


   1) change

              depvar=apply(sccs[,c("v473","v474","v475","v476")],1,sum)

      with

              depvar=sccs$v"your dep variable number goes here"  (remove quotes)

   2) change 
              dep_var=apply(sccs[,c("v473","v474","v475","v476")],1,sum),  #TOOK OUT v473

      with

              dep_var=sccs$v"your dep variable number goes here"  (remove quotes)

    you also want to change the following two lines as well:
    
   3)
      change
     
              name<-"how society values children"

      with
              name<-"your topic phrase goes here"  (this time keep the quotes)

   4)
       change

              alias<-"VLVDance"

       with
              alias<-"YOUR THREE INITIALS and TOPIC goes here without a space in between"   (again, keep the quotes)

Second set of changes:

  Now, you want to add the independent variables that you think may have and effect on your dependent variable. 
  There are already 100 or so independent variables defined in CreateModelValueSDWchildren2.R.
  First use them all and see how many of them are significant.


  1) Copy all the variable in the line that start with indep_vars<-c except the last one. 
     
    These are:
  
  "nonmatrel","famsize","exogamy","famsizeB","money","popdens","malesexag","ndrymonth","gath","hunt",
  "fish","anim","brideprice","nuclearfam","ncmallow","cultints","tree","roots","cereals","settype",
  "localjh","superjh","moralgods","segadlboys","plow","pigs","bovines","milk","agrlateboy",
  "fratgrpstr","Whyte577","Whyte580","Whyte584","Whyte585","Whyte595","Whyte602","Whyte615","Whyte620","Whyte626",
  "Whyte629","Whyte630","Whyte631","Whyte632","Whyte633","Whyte635","Paige657","Paige658","Paige659","Paige660",
  "Paige661","Paige662","fempower","migr","Sanday664","Sanday665","interperviol","Sanday667","Sanday668",
  "Sanday668","migr","Whyte718","Whyte719","Whyte720","Whyte721","Whyte722","Whyte723","Whyte724",
  "Whyte725","Rohner798","Rohner799","Rohner800","Rohner801","Rohner802","Rohner803","Rohner804","Rohner805",
  "Rohner806","Rohner807","Rohner808","Rohner809","Rohner810","Rohner811","Rohner812","Rohner813","foodtrade","dateobs",
  "rain","temp","ecorich","pctFemPolyg","femsubs","intwar","extwar","himilexp","AP1","AP2",
  "pathstress","war","intwarB","extwarB","foodscarc","sexratio","wagelabor"


  2) Paste them into the line that starts with 

     restrict_vars=c(

     make sure, your dependent variable is NOT one of them. If it is, remove it from the list. 


  3) Run the code and look at the output. You want to keep all the independent variables with a p value less than 0.125 and remove the rest.

  4) Go back to your line of 
     
     restrict_vars=c(

     this time remove all the variable names that were not significant in the first run of the code (i.e. those with p value BIGGER than 0.125) 

  This completes the second set of changes you should do. Remember to save your code under your wikipage. Have the code part under a title with A
  and the output with a title with B.


Third set of changes:

At this stage, you probably will want to add more independent variables to your model. So, in order to add more independent variables 
as opposed to using just those that were specified in the CreateModelValueSDWchildren2.R model we do the following:

1) find the number of the variable you want to add in the code book. 

2) add that variable into the list of variables at the beginning of your code and give a name to that variable. You do this in the following way:

   "your new independent variable name goes here"=sccs$v"the number of the independent variable goes here",   
   (remove quotes and make sure you have the COMA after the number unless you add your variable at the end of the list. 
    Basically, you want to make sure that if there are other independent variables being defined after the one you're adding
    then you want to put a coma to let R know that your independent variable is being followed by other independent variable)

3) Now go to the following two lines and add the new independent variable name in the lists of variables in these lines: 
  
   indep_vars<-c(


   and

   restrict_vars=c(

   make sure that when you're adding the name of the independent variable to these lists, it is in quotation marks and pay attention 
   to comas again depending on where in the list you're adding the independent variable name. You want a coma if you're adding it 
   in the middle of the list. 


4) Run the model with the newly added independent variable and see if it's significant. If it is keep it, if not remove it 
   from the list of restricted model ( the line that starts with restrict_vars=c( )

5) Add as many new independent variables as you think is necessary (I'd recommend as many as possible so long as you have
   a good justification as to why it may have an effect on your dependent variable-- articles and books supporting your claim
   is always a good thing but mere intuition can suffice sometimes as well)

This completes the last set of changes. Enjoy!

Not Working Code

1A Program CreateModelValueSDWchildren2.R

#--extract variables to be used from sccs, put in dataframe my_sccs--
setwd("C:/My Documents/sccs")
library(sccs)
data(sccs)                                  
depvar=sccs$v1711
my_sccs<-data.frame(
#--For this dep_var, we sum variables measuring how much a society values children--
#--can replace "sum" with "max"
dep_var=sccs$v1711,
socname=sccs$socname,
socID=sccs$"sccs#",
nonmatrel=sccs$v52,
famsize=sccs$v68,    # similar to v80
exogamy=sccs$v72,
famsizeB=sccs$v80,   # similar to v68
money=sccs$v155,
popdens=sccs$v156,
malesexag=sccs$v175,
ndrymonth=sccs$v196,
gath=sccs$v203,
hunt=sccs$v204,
fish=sccs$v205,
anim=sccs$v206,
brideprice=(sccs$v208==1)*1,
nuclearfam=(sccs$v210<=3)*1,
ncmallow=sccs$v227,
cultints=sccs$v232,
tree=(sccs$v233==4)*1,
roots=(sccs$v233==5)*1,
cereals=(sccs$v233==6)*1,
settype=sccs$v234,
localjh=sccs$v236-1,
superjh=sccs$v237,
moralgods=sccs$v238,
segadlboys=sccs$v242,
plow=(sccs$v243>1)*1,
pigs=(sccs$v244==2)*1,
bovines=(sccs$v244==7)*1,
milk=(sccs$v245>1)*1,
agrlateboy=sccs$v300,
### DEPVAR valchild=(sccs$v473+sccs$v474+sccs$v475+sccs$v476),
fratgrpstr=sccs$v570,
Whyte577=sccs$v577,
Whyte580=sccs$v580,
Whyte584=sccs$v583,
Whyte585=sccs$v584,
Whyte595=sccs$v594,
Whyte602=sccs$v602,
Whyte615=sccs$v615,
Whyte620=sccs$v620,
Whyte626=sccs$v626,
Whyte629=sccs$v629,
Whyte630=sccs$v630,
Whyte631=sccs$v631,
Whyte632=sccs$v632,
Whyte633=sccs$v633,
Whyte635=sccs$v635,
Paige657=sccs$v657,  # summed in v663#  Paige657 Paige658 Paige659 Paige660 Paige661 Paige662
Paige658=sccs$v658,  # summed in v663 #femproduceN/D=sccs$v658,
Paige659=sccs$v659,  # summed in v663
Paige660=sccs$v660,  # summed in v663
Paige661=sccs$v661,  # summed in v663
Paige662=sccs$v662,  # summed in v663
fempower=sccs$v663, # sum of v657-662
migr=(sccs$v677==2)*1,
Sanday664=sccs$v664,  # summed in v669
Sanday665=sccs$v665,  # summed in v669
#Sanday666=sccs$v666,  # summed in v669
interperviol=sccs$v666,   ###synonyms: violence / interviol / freintovio
Sanday667=sccs$v667,  # summed in v669
Sanday668=sccs$v668,  # summed in v669
Sanday669=sccs$v669, # sum of v664-668 
migr=(sccs$v677==2)*1, 
 #WHYTE Data Quality Whyte718 Whyte719 Whyte720 Whyte721 Whyte722 Whyte723 Whyte724 Whyte725
Whyte718=sccs$v718,
Whyte719=sccs$v719,
Whyte720=sccs$v720,
Whyte721=sccs$v721,
Whyte722=sccs$v722,
Whyte723=sccs$v723,
Whyte724=sccs$v724,
Whyte725=sccs$v725,
 #Rohner Data Quality Codes 
Rohner798=sccs$v798,
Rohner799=sccs$v799,
Rohner800=sccs$v800,
Rohner801=sccs$v801,
Rohner802=sccs$v802,
Rohner803=sccs$v803,
Rohner804=sccs$v804,
Rohner805=sccs$v805,
Rohner806=sccs$v806,
Rohner807=sccs$v807,
Rohner808=sccs$v808,
Rohner809=sccs$v809,
Rohner810=sccs$v810,
Rohner811=sccs$v811,
Rohner812=sccs$v812,
Rohner813=sccs$v813,
foodtrade=sccs$v819,
dateobs=sccs$v838,
rain=sccs$v854,
temp=sccs$v855,
ecorich=sccs$v857,
pctFemPolyg=sccs$v872,
femsubs=sccs$v890,
intwar=sccs$v891,    # similar to 1649
extwar=sccs$v892,    # similar to 1650
himilexp=(sccs$v899==1)*1,
AP1=sccs$v921,           ###agricultural potential 1
AP2=sccs$v928,           ###agricultural potential 2
pathstress=sccs$v1260,
war=sccs$v1648,     # overall -- sum of internal and external
intwarB=sccs$v1649,  # similar to v891
extwarB=sccs$v1650,  # similar to v892
foodscarc=sccs$v1685,
sexratio=1+(sccs$v1689>85)+(sccs$v1689>115),
wagelabor=sccs$v1732,
CVrain=sccs$v1914/sccs$v1913
)
indep_vars<-c(
"nonmatrel","famsize","exogamy","famsizeB","money","popdens","malesexag","ndrymonth","gath","hunt",
"fish","anim","brideprice","nuclearfam","ncmallow","cultints","tree","roots","cereals","settype",
"localjh","superjh","moralgods","segadlboys","plow","pigs","bovines","milk","agrlateboy",
"fratgrpstr","Whyte577","Whyte580","Whyte584","Whyte585","Whyte595","Whyte602","Whyte615","Whyte620","Whyte626",
"Whyte629","Whyte630","Whyte631","Whyte632","Whyte633","Whyte635","Paige657","Paige658","Paige659","Paige660",
"Paige661","Paige662","fempower","migr","Sanday664","Sanday665","interperviol","Sanday667","Sanday668",
"Sanday669","migr","Whyte718","Whyte719","Whyte720","Whyte721","Whyte722","Whyte723","Whyte724",
"Whyte725","Rohner798","Rohner799","Rohner800","Rohner801","Rohner802","Rohner803","Rohner804","Rohner805",
"Rohner806","Rohner807","Rohner808","Rohner809","Rohner810","Rohner811","Rohner812","Rohner813","foodtrade","dateobs",
"rain","temp","ecorich","pctFemPolyg","femsubs","intwar","extwar","himilexp","AP1","AP2",
"pathstress","war","intwarB","extwarB","foodscarc","sexratio","wagelabor","CVrain"
)
restrict_vars=c("cultints","roots","fish","exogamy","settype","femsubs")
library(foreign)
#--Read in two weight matrices--
Wll<-as.matrix(read.dta("./examples/data/langwm.dta")[,-1])
Wdd<-as.matrix(read.dta("./examples/data/dist25wm.dta")[,c(-1,-2,-189)])
load("./examples/data/vaux.Rdata",.GlobalEnv)
my_aux = vaux
row.names(my_aux)<-NULL
#--remove the society name field--
my_aux<-my_aux[,-28]
name<-"how society values children"
alias<-"VLVDance"
model=list(name=name,
           alias=alias,
           data=my_sccs,
           aux_data=my_aux,
           prox_list=list(language=Wll,distance=Wdd),
           dep_var="dep_var",
           indep_vars=indep_vars,
           restrict_vars=restrict_vars)
save(model,file=paste(alias,".Rdata",sep=""))
source("examples/src/run_model.R") #does for this model multiple imputation, two stage ols, saves to file to working directory.
lat<-sccs$v833.1
lon<-sccs$v833.2
plot(lon,lat, cex=.1)
ztxt=as.character(depvar) #above "my_sccs( " -- a new line inserted: depvar= as defined in my_sccs(
text(lon,lat,ztxt)
ols_stats$restrict_stats
ols_stats$r2
ols_stats$restrict_diagnostics
table(depvar)

https://www.facebook.com/tolga.oztan?fref=photo


new facts that might help to design your simulations

https://books.google.com/books?id=YtwSz2A12e8C&pg=PA335&lpg=PA335&dq=Ingalik+Iglulik&source=bl&ots=QP3ma5oRhH&sig=HetFXrS2V9CRHrGmL_smmYbDOSw&hl=en&sa=X&ei=t82_VPPINYSAgwS3zIGIBg&ved=0CCMQ6AEwAQ#v=onepage&q=Ingalik%20Iglulik&f=false

Has this new chapter by Chris Boehm. I can mail you the five missing pages from the google pages, but here is the important one: scan 4.jpeg attached

2nd table interesting: of those who solve conflicts by avoidance the following cases also have kin avoidance: Hadza, !Kung, Copper Estimo, Gilyak, Yahgan, Tiwi, Walbiri, Murngin

Anyway, this chapter gives a much broader understanding about cooperation and conflict resolution among Boehm's Late Pleistocene Adapted (LPA) foragers and you might be able to rethink these contexts when designing your simulations.

Our avoidance societies in WNAI and SCCS are from ethnographies of the last 300 years or so Whereas the LPA selection represents societies that might resemble those of 10,000 ypb years before present so your simulations might be designed to represent those different samples.

We never noticed that LRB also has a version of Boehm's conflict resolution

conpos. The posture of the particular group relative to the intensity of warfare within the region as coded in WAR.
Class=numeric; Type=ordinal; Number non-missing=336; Number of unique values=3
Freq    Value    Description
125    1    Generally defensive through avoidance rather than active defense
157    2    Tit-for-tat
54    3    Generally aggressive
3    NA    NA
and for kin avoidance
minlaw. The presence or absence of behavioral mother-in-law avoidance and other restrictions on behavior. Binary variable.
Class=numeric; Type=ordinal; Number non-missing=332; Number of unique values=2
Freq    Value    Description
161    1    Absent
171    2    Present
7    NA    NA

-- Doug White http://intersci.ss.uci.edu/wiki/index.php/DW_home